Polly: An Ecosystem that Enables Cell Metabolism Labs
Polly

Polly: An Ecosystem that Enables Cell Metabolism Labs

Kanika Sori
April 6, 2018

Discovery of a drug target requires expertise from different arms of a cell metabolism lab. A typical cell metabolism lab has a core biology facility that performs wet lab experiments, separate metabolomics and genomics core facilities, and a bioinformatics group to aid the core facilities. This is a case study of one such lab where Polly, a cloud-based data analytics platform, increased the throughput by bringing down hypothesis generation time from months to days.

Myra Johnson is a Research Assistant working with TargetA lab, a cell metabolism facility at Dale Medical School, that is interested in identifying drug targets for Diabetes type II patients.

Myra is primarily responsible for running experiments that her supervisor, Togo Stewart, designs. She prepares samples and runs them through the mass spec machine and collects raw data. She views and curates the peaks using El-Maven, a data processing engine for large-scale metabolomic experiments. Finally, she exports her table of peaks of interest to Polly and spots interesting metabolites across samples with Polly FirstView’s heatmap. She also looks at the intensity and fractional enrichment plots of metabolites of glycolysis cycle that pique her interest.

Next, she opens the project on Polly where she exports her processed data, and opens an iPython notebook to run the script that Sam D’Silva, a bioinformatician in her lab, wrote a year ago. This code allows Myra to normalise intensity values and re-plot intensity distributions. She converts the results into a knowledge repo that she shares with Togo to get his feedback. Polly enables Togo to review Myra’s work in one single ecosystem. Polly proves very fruitful in Myra’s and Togo’s discussion where Myra is quickly able to pull up data, peaks or plots Togo asked for. Togo is also able to dig deep into Myra’s data whenever he needs without her guiding him through her files and analysis.

Later, Myra extracts output data of the script and feeds it into IsoCorrect, an application on Polly, for natural abundance corrections. At the output screen, she hovers over suggested apps and uses PollyPhi Relative LCMS Dashboard to visualise key metabolites as canonical pathway visualisations. She exports these plots to a PowerPoint presentation that she can use in TargetA’s upcoming lab meeting.

Meanwhile, Togo re-evaluates Myra’s findings and generates a hypothesis – application of drug X2910 prevents the backflow of pyruvate to glucose and stimulates glycogen synthesis in mouse liver cells.

He is excited about it and asks Myra to repeat the experiment and confirm her findings before they present the hypothesis to Matt Pitt, a biologist from a core biology department. She takes the next two days to reproduce the analysis and confirms the hypothesis. She updates her presentation and is ready for the lab meeting scheduled for the next morning.

Intrigued? Book a Polly demo expert session now!

Subscribe to our Newsletter

Get the latest insights on Biomolecular data and ML

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Blog Categories